Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Color profile: Disabled Composite Default screen 6 Ecological Geography as a Framework for a Transition Toward Responsible Fishing Daniel Pauly, Reg Watson and Villy Christensen Fisheries Centre, University of British Columbia, Vancouver, Canada Abstract Meeting the widely expressed requirement that fisheries should somehow be managed on an ‘ecosystem basis’ implies that fisheries-relevant ecological processes, and the fisheries themselves, need to be documented in the form of maps. This allows recovery, in intuitive fashion, of at least some of the many dimensions of the complex ecosystems in which the fisheries are embedded. The implied transition, in fisheries science, from bivariate time series, to maps as major heuristic devices has a number of implications – some obvious, some less so – of which a number are here discussed and illustrated. Among the issues covered are: (i) the requirement for a consensus taxonomy of large marine ecosystems; (ii) the need to construct fisheries catch maps in the absence of positive records of what was caught where; (iii) the proper identification of one’s audience; and (iv) the mapping of marine protected areas and reserves. The seriousness of the fisheries crisis is emphasized, and the case is made that fisheries, if ever they are going to achieve some measure of sustainability – however defined – ultimately will have to be limited not only through the amount of effort they can effectively deploy, but also limited in space, leading to a change to the defaults under which fisheries operate, currently set such that all aquatic wildlife can be exploited, if under some restrictions. Introduction Fisheries worldwide are in serious trouble. There is perhaps no need to document this, but we shall still present a single graph, a time series of global marine catches, with and without the catches from China and of Peruvian anchoveta, which jointly mask the clear declining trend evident in the rest of the world’s fisheries and species (Fig. 6.1). This crisis has many aspects and proposed solutions, and one of the latter is the widely expressed requirement that fisheries management should somehow be put on an ecosystem basis – even though what this means is not yet very clear (NRC, 1999). The most common exhibits, in fisheries science, so far have usually been time series of key variables, e.g. catch, fishing mortality or spawning biomass. Such time series are usually hard to assemble, and their value increases with time (i.e. with the number of generations they encompass); hence the enormous value (in terms of both the costs they embody and the insights they led to) of the recruitment time series assembled by R.A. © 2003 by FAO. Responsible Fisheries in the Marine Ecosystem (eds M. Sinclair and G. Valdimarsson) 105 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:06 AM 87 Color profile: Disabled Composite Default screen 88 D. Pauly et al. Fig. 6.1. Trend in world marine fisheries catches as reported by FAO, with and without: (i) the catches from China (including Hong Kong and Macau, but excluding Taiwan) from Statistical Area 61 (Northwestern Pacific), which are massively over-reported (Watson et al., 2001); and (ii) the catches of Peruvian anchoveta, whose fluctuations largely reflect El Niño events. Removal of these two series unmasks a strong declining trend for the rest of the world’s fisheries, confirming the perception of widespread fisheries failures. Myers and collaborators (e.g. Myers et al., 1999). Indeed, the bivariate plots representing time series serve among fisheries scientists as a key heuristic device: we work hard to assemble them, show them to colleagues (see Fig. 6.1) and jointly ponder on their features, such as the amount of contrast they do or do not incorporate (Hilborn and Walters, 1992). Also, a huge number of methods have been developed to analyse such plots, ranging from time series analysis (Chatfield, 1984) and other statistical methods (regression, etc.), to simulation models, e.g. ECOSIM (see Walters et al., 1997), designed to produce (i.e. predict) time series. However, fisheries, embedded as they are in natural ecosystems, and relying as they do on natural fluxes of these systems, depend on the features of places. Thus, while we emphasize the variability of fisheries in time, we tend to lose track of their variability in space (Samb and Pauly, 2000). Indeed, we hardly use maps to discuss fisheries (except for tunas, see below). Maps, clearly, will be an important part of ecosystem-based management – though obviously they will not be all. Maps were crucial to the emergence of the modern world, as they catalogued the countries newly discovered by the European powers, and the best routes to their riches. The emergence of physical oceanography as a discipline of its own was also mediated by maps; indeed they are the currency that Commodore Matthew Maury (1806–1873), one of the founders of physical oceanography, used to ‘pay’ for the current, wind and depth observations that mariners sent him. Even our humble science of fisheries used maps to represent some of its newly acquired knowledge (Garstang, 1909; Fig. 6.2). Why we later neglected the device so successfully used by this and other pioneers of fisheries science to summarize their knowledge on the biology of North Sea or other fish need not be pursued here. What we can do, however, is to point out that the availability of powerful, PC-based geographical information system (GIS) technology makes it possible for maps to return to the central role that they had formerly in 106 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:07 AM Color profile: Disabled Composite Default screen Ecological Geography 89 Fig. 6.2. Schematic representation of the distribution of plaice (Pleuronectes platessa) in the North Sea, illustrating the key aspects of its life history (modified from Garstang, 1909). Contrast this information-rich map with the text that would be required to convey the same amount of information. summarizing knowledge in intuitive fashion, and hence this contribution. A number of issues will have to be sorted out, however, before fisheries maps become a routine tool in fisheries sciences: • • • • Ecosystem taxonomy and classification. Fisheries catch maps and related issues of scale. Using maps to reach new audiences. Maps and space-based fisheries management. Ecosystem Taxonomy and Classification Using fisheries maps for putting fisheries in an ecosystem context assumes that some agreement exists as to the definition and location of marine ecosystems. Indeed, without a prior definition of ecosystem boundaries, there is a real danger that the fisheries to be studied will themselves be used to define the boundaries of ecosystems pragmatically, as happened earlier with traditional biogeography, in which the distribution of the diverse groups mapped by specialists led to the definition of taxa-specific geographies, all mutually incompatible (see Ekman, 1967). Moreover, these taxa-specific geographies ended up being useless, even to those who had proposed them, due to the circularity of their definition. Thus, if the distribution of species within a given taxon defines a system of, say ‘provinces’, then features of these provinces diverging from what would be suggested (given the underlying physical 107 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:07 AM Color profile: Disabled Composite Default screen D. Pauly et al. 90 structure of the ocean) will not be identified, nor any resulting improvement to the system of provinces. The way out of this circularity is, of course, to use predefined ecosystem definitions (and boundaries), independent of (but hopefully with deep affinities to) the taxa or processes (here: fisheries) that are being mapped. There are at present three broad taxonomies-cum-classification systems representing the world ocean at scales below that defined by entire oceans, or their major basins, namely: • • The system of 18 Statistical Areas used by FAO to report global fisheries catches. The Large Marine Ecosystems defined by K. Sherman and collaborators. Box 6.1. • The system of four Biomes and 57 Biogeochemical Provinces described by Longhurst (1998) and presented in Box 6.1. The FAO’s system of statistical areas at present is the only device routinely used for breaking global catches into geographic space (Fig. 6.3). These 18 FAO areas are rather large, and have boundaries based largely on political considerations. Therefore, they cannot be used directly to put fisheries into an ecosystem context; however, they do provide some constraints for the rule-based construction of fisheries maps described in Box 6.2. Initially, large marine ecosystems (LMEs) were only what the three words in their name Biogeographical provinces. Until recently, a ‘geography of the sea’ did not exist that was suitable for describing, in standardized fashion, the distribution of all marine organisms, despite a history of oceanographic research starting with the Challenger Expedition (1872–1876). Numerous maps did exist in which this or that oceanographic parameter, or the distribution of a few organisms had been used to draw a map of some sort (see, for example, Ekman, 1967). However, no tests were conducted of the ability of these maps to predict distributions other than those from which they were derived: circularity reigned supreme. Reasons for this are easy to imagine, from the excessive preoccupation of various specialists with their favorite taxonomic groups, to the absence, before the computer revolution, of analytic tools that were up to the task. However, the real reason is probably that developing a truly synoptic vision of the ocean was impossible before the advent of satellite-based oceanography. Satellites cannot ‘see’ very deep into the sea. However, what satellites do see is the very stuff that generates fundamental differences between ocean provinces: sea surface temperatures and their seasonal fluctuations, and pigments such as chlorophyll, and their fluctuations. Marine systems differ from terrestrial systems in that their productivity is essentially a function of nutrient inputs to the illuminated layers. This gives a structuring role to the physical processes that enrich surface waters with nutrients from deeper layers, such as wind-induced mixing, fronts, upwellings, etc. (Longhurst, 1995). Thus, the location, duration and amplitude of deep nutrient inputs into different oceanic regions – as reflected in their chlorophyll standing stocks – largely define the upper trophic level biomasses and fluxes that can be maintained in these regions. This is the reason why satellite images reflect fundamental features of the ocean, while maps based on the distribution of various organisms – even ‘indicator’ organisms – can only reflect second-order phenomena. T. Platt, S. Sathyadranath and A.R. Longhurst are among the first to have realized this, and thus their stratification of the ocean, and the estimates of global primary production based thereon, are far superior to earlier attempts. The system of biomes and biogeochemical provinces defined in the process was refined further in a book by Longhurst (1998), the review of which (Pauly, 1999) provides the basis for this box. One interesting aspect of this stratification (or classification) is that the biogeochemical provinces (BGCPs) in the ‘coastal biome’ thus defined largely overlap with the large marine ecosystems (LMEs) of Sherman and collaborators (Pauly et al., 2000; see Figs 6.4 and 6.5a and b). This correspondence should make it possible to integrate in a common framework the vast amount of geo-referenced information on marine biological processes that is now available, and finally to make widely available to practitioners the data that so many of them still claim we do not have. 108 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:08 AM Color profile: Disabled Composite Default screen Ecological Geography 91 Fig. 6.3. System of 18 statistical areas used by FAO to report on global marine fisheries statistics. Note the large size of these areas, encompassing very different ecosystems and faunas. Box 6.2. Construction of fisheries maps (by R. Watson). The records used for constructing detailed catch maps (here: by half a degree lat./long. cells) for an ocean basin (or worldwide for global maps) for a given year are based either; (i) exclusively on FAO statistics on the countries fishing in that basin, or on the FAO global statistics; (ii) on FAO statistics complemented with time series of discards, estimates of illegal catches, etc.; or (iii) statistics that substitute for those of FAO, e.g. International Council for the Exploration of the Sea statistics in the Northeastern Atlantic, complemented as in (ii) or not (A in Fig. 6.6a). These are processed as a set of database records by first disaggregating the statistics for the generalized group into records at lower taxonomic levels (B in Fig. 6.6a), as necessary for many countries where the reported catch composition is very aggregated, such as China, and using a catch composition interpolated from that of immediate neighbours with detailed statistics (here Taiwan and South Korea). Then, each taxon represented in a landing record is looked up in a database of species-specific spatial distributions that identifies the subset of spatial cells of the world’s oceans from which the catch record in question could originate. The country reporting (fishing) is then looked up in; (i) a database of fishing access agreements (updated from Anon. 1998); and (ii) a database identifying the exclusive economic zones (EEZs) of the world’s countries (see Table 6.1), which jointly identify the spatial cells that are available for that country to fish in (including the EEZ of other countries for which arrangements exist). The FAO area that the statistic was reported from is also used to identify a set of spatial cells from which the catch may originate. These sets of spatial cells are then compared and if there are no overlapping cells the landing is not allocated and an ‘error report’ is logged (see ‘no’ in Fig. 6.6b). Otherwise, the reported landing is assigned among overlapping cells in proportion to their areas. Thus, landing rates (t km−2 year−1) are accumulated in each cell as each record is processed (currently, we are able to allocate > 95% of the world catch to cells; the remainder reflects error reports whose resolution we expect to contribute to cleaning up the underlying databases. In this way, a grid map of landing rates is built up as each landing record is processed (D in Fig. 6.6a and b). Though each record is processed for the taxonomic level it is reported at (after disaggregation), the results can be reassembled into larger groups as required, e.g. for statistical models (E in Fig. 6.6a and b). Alternatively, the taxon- and cell-specific catch records can be multiplied by its corresponding market price, yielding maps of catch value, a new product for which the Sea Around Us Project envisages a large range of uses. 109 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:08 AM Color profile: Disabled Composite Default screen 92 D. Pauly et al. imply, namely marine ecosystems defined so as to cover a large area (200,000 km2 or more). Gradually, however, and mainly due to the work of K. Sherman and collaborators, LMEs became restricted to an explicit list of 50 coastal entities (Fig. 6.4), with a dozen recently added (see www.edc.uri.edu/lme/default. htm), broadly defined by physical features (presence of shelves, coastal currents, fronts, etc.), and documented in a growing number of books (listed in www.edc.uri.edu/lme/ publications.htm). One major conservation-oriented nongovernmental organization (NGO), the World Conservation Union (IUCN), has endorsed the concept, with the intention of using it for reporting on marine biodiversity. Similarly, FishBase, the global database of fish, has linked all species of marine fishes in the world (~ 15,000 species) with the LME in which they occur (see Table 6.1). One important features of LMEs is that, while covering only 18% of the world’s oceans, they accounted for 75% of the world’s fisheries catches in 1999. These figures, based on the 50 LMEs listed in Sherman and Duda (1999), will increase when recalculated based on the 62 LMEs in Fig. 6.4. The most rigorous division of the world ocean, at least in terms of biological oceanography, is, however, the system described by Longhurst (1998), based on Platt and Sathyendranath (1988), Sathyendranath et al. (1995) and Longhurst (1995) (see Box 6.1 and Fig. 6.5a and b). At the highest level, this hierarchical classification is based on a division of the world ocean into four biomes. In the Polar biome, covering only 6% of the world ocean, vertical density structure is determined very largely by low-salinity water derived from ice-melt each spring. In the Westerlies biome, between the Polar fronts and the Subtropical Convergence, large seasonal differences in mixed-layer depth are forced by seasonality in surface irradiance and wind stress, inducing strong seasonality of biological processes, characteristically including a spring bloom of phytoplankton. The Trade-wind biome lies across the equatorial regions, between the boreal and austral Subtropical convergences, where low values for the Coriolis parameter, a strong density gradient across the permanent pycnocline and weak seasonality in both wind stress and surface irradiance, result in relatively uniform levels of primary production throughout the year. Finally, the Coastal Boundary biome is composed of the continental shelves and the adjacent slopes, i.e. from the coastlines to the oceanographic front usually found at the shelf edge (Pauly et al., 2000). Next, the biomes are subdivided into 57 biogeochemical provinces (BGCPs), defined by satellite imagery and physical oceanography. Each BGCP is characterized by a distinct regime of physically driven water mixing, leading to a distinct pattern of (seasonal) supply of nutrients to the euphotic zone, and hence primary production (Longhurst et al., 1995; Longhurst, 1998; see also Table 6.1). In this scheme, the BGCPs comprising coastal biomes largely overlap with the area covered by the LME mentioned above, and hence the suggestion of a consensus system in Pauly et al. (2000), currently being implemented through a collaboration between members of various teams represented by the authors of the consensus statement. A further advantage of the consensus approach implied by the structure provided by LME/BGCP is that it leads to emphasizing benthic–pelagic coupling, as a single set of ecosystems is proposed for the neritic (shelf) areas of the world. This is appropriate, as it counters the misguided tendency to separate the pelagic and benthic realms, which leads to ecosystem representations that are exceedingly ‘open,’ and in which benthic–pelagic coupling must be represented explicitly (and thus quantified). Rather, benthic–pelagic coupling should be allowed to appear as an emergent property of neritic food webs, as will occur when one’s ecosystem representation includes predators feeding both on benthic and pelagic organisms, and detritus (e.g. marine snow) that is consumed both while sinking, and after it has sedimented. Conversely, there is no need for benthic–pelagic coupling in representations of open ocean systems, where the pelagic (sub)system is largely independent of benthic processes, and can be modelled as such; see, for example, Kitchell et al. (1999). Another reason to be wary of uncoupling the benthic and pelagic components of neritic 110 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:09 AM Color profile: Disabled Composite Default screen Ecological Geography Fig. 6.4. System of large marine ecosystems (LMEs) identified by K. Sherman and collaborators. This maps includes 12 recently defined LMEs (notably around Australia; see Table 6.1). 93 111 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:11 AM Color profile: Disabled Composite Default screen 94 D. Pauly et al. Fig. 6.5. (a) The four ‘biomes’ in the global ocean stratification of A.R. Longhurst and colleagues (Polar, Westerlies, Trade-Winds and Coastal Boundary). Note their overall match with a global climate map (insert, from Anon., 1991). (b) Biogeochemical provinces (BGCPs) in the system of A.R. Longhurst and collaborators. Note that each BGCP fits into one of the four biomes in (a), thus allowing for a nested hierarchy of comparable ecosystems. systems is that fishing itself tends to turn ecosystems dominated by benthic organisms (in terms of biomass or species numbers) into systems dominated by (small) pelagics and planktonic organisms. This feature, initially documented as a response to the stress generated by the combined effects of pollution and overfishing in the North American Great Lakes, has now been shown capable of being induced by fishing alone – at least in principle (see Parsons, 1996). Broadly speaking, this would be due to trophic cascades, wherein fewer piscivores → more small pelagics → fewer zooplankton → more phytoplankton. Such indirect effects are very hard to identify in practice, given the contribution of terrigenous fertilizers in regions plagued by algal blooms, such as the northern Gulf of Mexico (Turner and Rabalais, 1994) or the inner Gulf of Thailand (Piyakarnchana, 112 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:13 AM Color profile: Disabled Composite Default screen Ecological Geography 1999). However, the possibility that such effects can occur provides a good additional reason for constructing models of neritic systems that integrate the entire water column, and not only their benthic or pelagic components. Construction of Catch Maps and Issues of Scale The scope of fisheries science, and of the related components of marine biology, traditionally has been defined by the scale of the fisheries studied (Pauly and Pitcher, 2000), which may range from a few square kilometres or even less (e.g. in the case of fisheries for sessile invertebrates) to thousands of square kilometres in the case of high sea fisheries. However, basin-level analyses are rare, except for tuna fisheries (for which, incidentally, mapping frequently is used; see below). Over 75% of fisheries landings (in value) are consumed in countries other than those owning the exclusive economic zone (EEZ) in which these landings were realized (based on FAO, 2000). In contrast, only 4–5% of the rice grown in the world is traded internationally Table 6.1. 95 (Maclean, 1997). This, by itself, provides a rather good reason why global fisheries maps are appropriate to our times – not to mention the need to quantify the global impacts of fisheries on marine ecosystems. One objection frequently heard relating to the feasibility of large-scale fisheries maps is the absence of suitable data, widely understood to consist of positive records of where some fishing unit may have caught, at a certain time, a certain quantity of fish (as plotted in tuna atlases) (Fonteneau, 1997; see Table 6.1 for FAO atlas). Such records, usually supplied by the industry, or costly observer programmes, are indeed rather scarce and, when available, are either presented at very coarse scales (e.g. in 5° scale for the FAO tuna atlas, to mask small-scale patterns with the high concentrations so dear to the industry) or pertain to small areas, and the catch of a limited set of gear. Constructing global fisheries maps from such data, i.e. from the ‘bottom-up’, does indeed seem unfeasible. However, such ‘positive’ records are not required to construct fisheries catch maps. These can also be constructed from the geographic range of the exploited taxa, and constraints on which parts of that range led to the reported catches, i.e. from ‘negative’ records as it were. The Databases (on-line or CD-ROM) used for the construction of fisheries catch maps. Data type Organization URL Fisheries landings Tuna and billfish landings by 5° cells Taxonomy for all species, and ranges for many Distribution of commercial fish and invertebrates Physical ocean data (depths, temp. etc.) Primary productivity FAO FAO www.fao.org/fi/statist/FISOFT/FISHPLUS.asp www.fao.org/fi/atlas/tunabill/english/home.htm FishBase www.fishbase.org FAO www.fao.org/fi/sidp/default.htm NOAA www.ngdc.noaa.gov/mgg/global/global.htm Coral reefs Seamounts Sea ice extent Exclusive economic zones Fishing agreements www.me.sai.jrc.it/me-website/contents/shared_utilities/ frames/index_windows.htm www.reefbase.org ReefBase www.ngdc.noaa.gov/mgg/global/global.htm NOAA nsidc.org/index.html Univ. Colorado Veridan Information www.maritimeboundaries.com/main.htm Solutions (see FAO, 1998) Contact FAO FAO JRC of the EU 113 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:13 AM Color profile: Disabled Composite Default screen 96 D. Pauly et al. procedure used for this is presented in Box 6.2 and Fig. 6.6a and b, while Table 6.1 indicates sources of information on the key databases used in the process. This approach works straightforwardly at larger scales (FAO areas, biomes, ocean basins or global), but not at smaller scales, wholly comprised within the geographic range of a number of resources species, where positive knowledge on fleet operations is required. Issues of scale also show up when dealing with the definition of an ecosystem. Indeed, such issues are implicit in its definition as an ‘area where a set of species interact in characteristic fashion, and generate among them biomass flows that are stronger than the flows linking that area to adjacent ones’ (Pauly and Froese, 2001). This definition applies to the large scale inherent in the LMEs and BGCPs presented above, but also the small scale (a few hundred metres), where organisms interact to form the food webs that characterize coral reefs, or small lagoon or estuarine systems. Using maps to reach new audiences It is not the fishing industry that is asking for fisheries management to be put on an ecosystem basis, but politicians, pressed by conservation-orientated NGOs, themselves expressing public unease about the way marine ecosystems, and especially their more charismatic components (marine mammals, turtles and birds), are being affected by fishing. Thus, progress in putting fisheries on an ecosystem basis will have to be reported to that audience – quite a change for fisheries scientists accustomed to generate total allowable catches (TACs), communicated to highlevel bureaucrats by their superiors, fought by industry representatives, then applied or not to contain a fishery on the ground. It may be argued that the public at large will not understand the message conveyed by maps of fisheries and their ecosystem impacts. Yet, every day, a large part of the population, in most countries of the world, 114 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:14 AM Color profile: Disabled Composite Default screen Ecological Geography watch television weather programmes, and understand the sophisticated weather maps presented therein, although they are based on millions of data points analysed in quasi-real time by supercomputers, and combine dynamic displays of temperature, atmospheric pressure, cloud cover, risk of precipitation, etc. The public has been able to learn the ‘language’ of weather maps because: (i) it matters and (ii) visual displays presented in intuitive fashion can convey far more information than a text that is read or heard (Tufte, 1983). Thus, engaging the public and our political representatives in debates about the state of fisheries resources, and about alternative approaches to their utilization and long-term sustainability, should be possible, if we use the proper format for conveying that 97 information. We believe that maps provide that format, and we have documented this with maps illustrating the decline of piscivorous fishes in the North Atlantic (see Box 6.3). It is our belief that without such engagement with the public, the fisheries sector, including the fisheries science that studies it, and the largely captive regulatory agencies that ‘manage’ the resources, will not be able to halt the decline illustrated by the biomass maps described in Box 6.3. Maps and space-based fisheries management What is striking when examining the catch or biomass maps (Fig. 6.7) is that all show Fig. 6.6. (a) (Opposite) Schematic representation of algorithm for construction of catch maps in the absence of positive, georeferenced catch records: the algorithm is initiated (in A) with global catch statistics from FAO, or from regional or national sources; its output is cells to which catches have been assigned (see also Fig. 6.5b and Box 6.2). (b) Schematic representation of algorithm for construction of catch maps in the absence of positive records. A catch record (from FAO database or other source) is evaluated via three criteria (what taxon, by which country, in which FAO area), and can be assigned only when one or more cells meet these criteria. Over 95% of the world catches from FAO can be assigned straightforwardly in this fashion; the remainder providing pointers to corrections of the assignment rules (see also (a) and Box 6.2). 115 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:15 AM Color profile: Disabled Composite Default screen D. Pauly et al. 98 Box 6.3. Examples of fisheries ‘weather’ maps for the North Atlantic (by V. Christensen). The North Atlantic is one of the best studied marine areas of the world, which is not surprising since the marine sciences emerged largely along its shores, from 150–100 years ago. As a consequence, and contrary to the stubborn beliefs, by various colleagues, in a widespread ‘lack of data,’ abundant data sets exist which can be used to evaluate the impact of fisheries on the North Atlantic ecosystem. This is one of the goals of the of the Sea Around Us Project (SAUP; see www.fisheries.ubc.ca/projects/SAUP and Pauly and Pitcher, 2000). However, these data sets do not have the form required for analysis. (Note though, that this is always true: it is the analytical process itself which determines the format data should have.) After opting to present the SAUP analyses in form of ‘weather maps,’ a two-step approach was used for their construction: 1. Construction of mass-balance food web models for all major shelf areas, and a representative subset of oceanic areas to quantify biomasses at different trophic levels, and for different periods. 2. Extension, using a statistical model, of the biomass estimates in (1) to the entire North Atlantic, and the period from 1950 to 2000. Item (1) relied on 17 models constructed by a vast number of authors, most associated with national research institutions in the countries concerned (for details, see contributions in Guénette et al., 2001). Importantly, these models included data-rich representations of the North Sea in 1880, and the Newfoundland shelf in 1900, both implying higher biomass of predatory fishes than in the corresponding 1980s models, and several other model pairs, contrasting present biomass with those in the 1970s or 1960s. The fish biomasses in these models, all constructed using the ECOPATH WITH ECOSIM software (EwE), were put on a spatial basis using the Ecospace routine of EwE (Walters et al., 1999), using the same half-degree spatial cells also used to construct catch maps (see Box 6.2). Item (2) then consisted of identifying a general linear model of the form Biomasstyc = f (catchyc; year and physical attribute of half-cell) wherein the subscripts are t = trophic level, y = year and c = catch in each half-degree cell (mapped as presented in Box 6.2), and where the (bio-)physical attribute (assumed invariable in time) of each cell include mean depth and temperature, primary production, ice cover and other properties (see Christensen et al., 2001, for details). Examples of the maps thus generated, which indicate a strong decline of predatory fish biomass from 1950 to the present, indicative of ‘fishing down marine food webs’ (Pauly et al., 1998, 2001) are available on-line (www.fisheries.ubc.ca/projects/SAUP). These maps highlight processes that are rather worrisome, and none of the persons (both specialists and laypersons) to whom they have been shown has failed to perceive their analogy to weather maps, and to very bad weather developing over the North Atlantic. the same trend, at least for the North Atlantic, the only ocean basin we have examined so far in some detail. This is not surprising, given that we did not include local components in the algorithms used to generate these maps. The point, though, is that at the scale we were working (with ~ 21,000 pixels of half a degree latitude and longitude), there were no marine reserve, or other refugia (with biomass trends different from the general downward trend for the North Atlantic as a whole) to consider. Put differently: there are – at the scale of our pixels, appropriate to represent the distribution ranges of all but small intertidal species – no areas of the North Atlantic where fishes are not exposed to nets and other gear designed for, and extremely effective at, catching them. The deleterious effects of the sort of default setting implied here (i.e. that fish can be exploited anywhere, unless regulations state otherwise, rather than the converse) are discussed by Walters (1998). Put as a map, contrasting areas with fishing (say red) versus areas without any fishing, this would imply a single colour for the entire North Atlantic, without any green, or other shades (Fig. 6.8). We are 116 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:15 AM Color profile: Disabled Composite Default screen Ecological Geography 99 Fig. 6.7. Catch map for the North Atlantic (here: 1990s), constructed as explained in Box 6.2 and Fig. 6.6a and b, with darker shades indicating higher catch rates, in t km−2 year−1. Colour versions of this and similar maps for other periods and areas are available on-line; see Box 6.3. Fig. 6.8. Map of the North Atlantic, with dark identifying those areas where fishing is allowed, and fish killed, and light the areas where no fishing is permitted, thus allowing the resource to recover. Unfortunately, there are no light areas to be seen at the scale of half-degree cells. (The authors welcome corrections that would identify recently created refugia.) confident that, as for the weather maps presented above, the meaning of this monochromatic map will be widely understood by lay audiences. Figure 6.8 would be less dire had it been designed to illustrate the extent of protection for coral reefs and other sensitive coastal systems (e.g. kelp beds). For these, the idea of area-based protection is well accepted, and a number of (mainly small) marine reserves have been created. Here, high-resolution maps are understood as playing the key role in defining the terms of the debate between different 117 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Thursday, January 16, 2003 2:16:41 PM Color profile: Disabled Composite Default screen D. Pauly et al. 100 stakeholder groups, namely local residents, small-scale fishers, dive-resort operators, environmental NGOs, etc. In contrast, it may take a while for the management of commercial fisheries to rely on area-based protection as its key tool. When this happens, maps will be there to help us see where we are, and where we should be going. Acknowledgements We thank Mr A. Gelchu for shape files of fish distribution, Dr U.R. Sumaila for estimating the fraction of the world’s fish catch value that is exported, and our colleagues at FAO for fruitful exchanges. We also thank the Pew Charitable Trusts for their support of the Sea Around Us Project. References Anon. (1991) Bartolomew Illustrated World Atlas. Harper Collin, Edinburgh. Chatfield, C. (1984) The Analysis of Time Series: an Introduction. Chapman and Hall, London. Christensen, V., Guénette, S., Heymans, S., Walters, C.J., Watson, R., Zeller, D. and Pauly, D. (2001) Estimating fish abundance of the North Atlantic, 1950 to 2000. In: Guénette, S. Christensen, V. and Pauly, D. (eds) Fisheries Impacts on North Atlantic Ecosystems: Models and Analyses. Fisheries Centre Research Reports 9(4), 1–25. www.saup.fisheries.ubc.ca/report/ report.htm Ekman, S. (1967) Zoogeography of the Sea. Sidgwick and Jackson, London. FAO (1998) FAO’s fisheries agreements register (FARISIS). Committee on Fisheries, 23rd Session, Rome, Italy, 15–19 February 1999 (COFI/99/Inf.9 E). FAO (2000) Fisheries trade flow (1995–1997). FAO Fisheries Circular, No. 961. Fonteneau, A. (1997) Atlas of Tropical Tuna Fisheries. Edition ORSTOM, Paris. Froese, R. and Pauly, D. (eds) FishBase 2000: Concepts, Design and Data Sources. Los Baños, Philippines. Garstang, W. (1909) The distribution of the plaice in the North Sea, Skagerrak and Kattegat, according to size, age and frequency. Report of the trawling investigations of the research streamers from October 1902 to July 1907. Rapports et Procès-Verbaux des Réunions du Counseil Permanent International pour l’Exploration de la Mer 11, 136–138. Hilborn, R. and Walters, C.J. (1992) Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty. Chapman and Hall, New York. Kitchell, J.F., Boggs, C.H., He Xi and Walters, C.J. (1999) Keystone predators in the Central Pacific. In: Ecosystem Approaches for Fisheries Management. Alaska Sea Grant College Program, AK-SG-9901, pp. 665–684. Longhurst, A.R. (1995) Seasonal cycles of pelagic production and consumption. Progress in Oceanography 36, 77–167. Longhurst, A.R. (1998) Ecological Geography of the Sea. Academic Press, San Diego. Longhurst, A.R., Sathyendranath, S.A., Platt, T. and Caverhill, C.M. (1995) An estimate of global primary production in the ocean from satellite radiometer data. Journal of Plankton Research 17, 1245–1271. Maclean, J. (ed.) Rice Almanac. IRRI, Los Baños, Philippines. Myers, R.A., Bowen, K.G. and Barrowman, N.J. (1999) The maximum reproductive rate of fish at low population sizes. Canadian Journal of Fisheries and Aquatic Sciences 56, 2404–2419. NRC (National Research Council) (1999) Sustaining Marine Fisheries. National Academy Press, Washington, DC. Parsons, T.R. (1996) The impact of industrial fisheries on the trophic structure of marine ecosystems In: Polis, G.A. and Winnemiller, K.D. (eds) Food Webs: Integration of Patterns and Dynamics. Chapman and Hall, New York, pp. 352–357. Pauly, D. (1999) Review of A. Longhurst’s ‘Ecological Geography of the Sea’. TREE 14(2), 118. Pauly, D. and Froese, R. (2001) Fish stocks. In: Levin, S. (ed.) Encyclopedia of Biodiversity, Vol. 2. Academic Press, San Diego, pp. 801–814. Pauly, D. and Pitcher, T.J. (2000) Assessment and mitigation of fisheries impacts on marine ecosystems: a multidisciplinary approach for basin-scale inferences, applied to the North Atlantic. In: Pauly, D. and Pitcher, T.J. (eds) Methods for Evaluating the Impacts of Fisheries on North Atlantic Ecosystems. Fisheries Centre Research Reports 8(2), pp. 1–12. www.saup.fisheries.ubc.ca/report/report. htm Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. and Torres, F.C., Jr. (1998) Fishing down marine food webs. Science 279, 860–863. 118 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:16 AM Color profile: Disabled Composite Default screen Ecological Geography Pauly, D., Christensen, V., Froese, R., Longhurst, A., Platt, T., Sathyendranath, S., Sherman, K. and Watson, R. (2000) Mapping fisheries onto marine ecosystems: a proposal for a consensus approach for regional, oceanic and global integration. In: Pauly, D. and Pitcher, T.J. (eds) Methods for Evaluating the Impacts of Fisheries on North Atlantic Ecosystems. Fisheries Centre Research Reports 8(2), pp. 13–22. www.saup.fisheries.ubc.ca/report/report. htm Pauly, D., Palomares, M.L., Froese, R., Sa-a, P., Vakily, M., Preikshot, D. and Wallace, S. (2001) Fishing down Canadian aquatic food webs. Canadian Journal of Fisheries and Aquatic Sciences 58, 51–62. Piyakarnchana, T. (1999) Changing state and health of the Gulf of Thailand large marine ecosystem. In: Sherman, K. and Qisheng Tang (eds) Large Marine Ecosystems: Assessment Sustainability and Management. Blackwell Science, Malden, UK, pp. 240–250. Platt, T. and Sathyendranath, S.A. (1988) Oceanic primary production: estimation by remote sensing at local and regional scales. Science 241, 1613–1620. Samb, B. and Pauly, D. (2000) On ‘variability’ as a sampling artefact: the case of Sardinella in north-western Africa. Fish and Fisheries 1, 206–210. Sathyendranath, S.A., Longhurst, A.R., Caverhill, C.M. and Platt, T. (1995) Regionally and seasonally differentiated primary production 101 in the North Atlantic. Deep-Sea Research 42, 1773–1802. Sherman, K. and Duda, A.M. (1999) An ecosystem approach to global assessment and management of coastal waters. Marine Ecology Progress Series 190, 271–287. Tufte, E.R. (1983) The Visual Display of Quantitative Information. Graphic Press, Cheshire, Connecticut. Turner, R.E. and Rabalais, N.N. (1994) Coastal eutrophication near the Mississippi river delta. Nature 368, 619–621. Walters, C.J. (1998) Designing fisheries management systems that do not depend upon accurate stock assessments. In: Pitcher, T.J., Pauly, D. and Hart, P. (eds) Reinventing Fisheries Management. Fish and Fisheries 23. Kluwer Academic, Dordecht, The Netherlands, pp. 279–288. Walters, C.J., Christensen, V. and Pauly, D. (1997) Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Reviews in Fish Biology and Fisheries 7(2), 139–172. Walters, C.J., Pauly, D. and Christensen, V. (1999) ECOSPACE: prediction of mesoscale spatial patterns in trophic relationships of exploited ecosystems, with emphasis on the impacts of marine protected areas. Ecosystems 2, 539–554. Watson, R., Pang, L. and Pauly, D. (2001) The marine fisheries of China: development and reported catches. Fisheries Centre Research Reports 8(2). www.saup.fisheries.ubc.ca/ report/report.htm 119 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:17 AM Color profile: Disabled Composite Default screen 120 Z:\Customer\CABI\A4396 - Sinclair\A4469 - Sinclair - First Revise #A.vp Wednesday, January 08, 2003 10:38:17 AM